TW442759B - Improving the portability of digital images - Google Patents

Abstract

Techniques for storing and translating digital images of different native formats in a way that makes it unnecessary for independent manufactures of imaging devices such as digital cameras to agree on factors related to production/capture and processing of the images. Image data in a native format and an associated method are combined as part of an image object, within the meaning of classical object-oriented technology, and transferred to the host system. An abstract machine, such as a virtual machine, in the host system acts as a virtual image processor and executes the method to obtain image data in a common format. The method in each image object operates on its corresponding image data to yield an image in a common format.

Description

V. Description of the Invention (l) '^ —----_____ BACKGROUND OF THE INVENTION Field of the Invention In general, the present invention relates to digitally generated images, which can be in different native formats, and to be related to the technology of packaging numbers. Dong Qi's related technology, digital imaging devices such as digital cameras, which use a common format translation, have gradually chosen to integrate personal computers (p. Industry and photography industry integration). The unprecedented image idealization is suitable for the combination of personal computers and brains. It seems complete. But in fact, this item combines m with the current practice of the digital photography industry of the personal computer system == hindrance, j. This point cannot be changed .. Various novel camera codes: Use this camera to work with the image, the bite is specially set from the camera of a different manufacturer; $ = type digital camera or come. Stored outside the camera, each y & how to deal with digital images with little or no digital images 1Λ Λ For example ::: is used as a view between the device and the host. ^ # ^ CTwa.n) ^, 〇 #,-Silly. ^ The corresponding program provides a standard interface for the imaging group class 'Γ 人 胜 吴 \' to make: the group translation image is captured by the imaging device's native format to capture digital digits. 02 common format. C process company definitions in these common formats. Device-independent bitmaps ... such as Microsoft

Page 5 ^ 42759 V. Description of the invention (2) Different types of digital devices and local GTE modules need to transfer image storage data to digital cameras. These modules generate security risks. The solution is to hope that there is a format to store the digital data in a common format. The drive is loaded. Overview Cameras are often used and stored with digital hosts outside of each new host. The reason is that direct communication with the host, such as viruses, is not really possible at a specific host. The technology is used to visualize the system with subjective viewing or processing. It ’s different from the ‘Operation Department’. But the real digital phase operation system can spread the sharing of the processor. Allows you to store images in a digital format. The specific combination of the designated combination of the (0S) < 'two makes the special temperature module can be used together with the limited memory resources of such module memory. Such as the system interaction, so the camera into the host system. Finally, the Twenty specificity can be written, so that the same type of imaging device can communicate with each other, so it is allowed to the host without the standby device-specific special device " specified in a specific example about an imaging device having An image sensing object is used to generate sensor material and memory for storage-image object. Video, which has sensor-related image data, and a method for obtaining translated image data based on the image material by an extraction machine. The brief summary of the specific examples of the invention of Γ, as well as other features and advantages of I, will be clearly explained by the detailed description, the scope of patent application, and the appendix. Brief description of the diagram, which is the overall system of an imaging system according to the present invention, a specific example

442759 V. Description of the invention (3)-Figure 2 shows the steps of another specific commonality according to the present invention. Private Example Ordering to Improve Digital Imaging Figure 3 shows a host processing system, including configuration software and hardware. # Other-specific according to the present invention Fig. 4 shows a part of an imaging device including software and hardware components configured according to another specific embodiment of the present invention. Detailed description As briefly explained above, the present invention provides the ability for a camera manufacturer to select its own local device such as digital image data, while at the same time depending on the host's specific knowledge of the specific imaging device in which it is arranged and stored. The image solution uses the r: part of the image. In this way, each image uses one image: two storages as the object of the object, as the typical meaning of the object oriented software = no, but the package contains image data and related methods. This = ::-part. Image Objects The wisdom required to pan to a common image file format. ^ Is a balance of specificity that supports the native format of data from native formats. Method Definition Associated Image Explained by the chief executive who performed the imaging method. All systems are intact, including hardware and operations; all are :: pair :: two: processing It is distributed independently of the imaging device. This post is over for development as well as for example virtual machines. Due to the fact that the environment is performed by the virtual object, there is no difference between the common format and the image data used by the ^. ^ Different and new imaging is issued. 2 The independent method developed by J f To identify and process the target ^ total ^ = formula can also eliminate the need to develop explicit requirements. Image objects can be imaged :: Same as image file format. Device manufacturers are exempted from having to decide what is storage.

Page 7

442769

Image. Application 1 24 can be set to further translate each image into one of a variety of other audit formats, including collaborative photography expert group format 0 (JPEG), graphics interchange format (GIF), interchange file format (IFF), and tagged images File format (TI FF). For the sake of convenience, the following discussion only refers to the most rationale other than the storage of imaging and imaging devices. This solution is a special format, or decides which is the best use. Changes in the format of the image file are changed. The reason is that the change frequency of virtual machines and / or software is much lower. Good frequency may be higher than the imaging device or the application circle 1. The overall system view of the various figurative objects of the present invention being formed or stored in different ones, of which the shadow device 110 is subsequently transferred to the image devices 1 and 104b and stored. Or 108c includes at least—information 呷 俨. Discord 102. Each object 108a ’108b is a local device of a specific imaging device type, which is used to capture the image data. 104a may be configured as a scanner to be able to be used.

1 ^ ^ 4 J House Health-Object 1 08a, which represents the early image paid using this G position. On the contrary,-a single object 108b can be generated, and a digital camera is configured to be set to 10. There are two data parts 1 1 4 ratio and 11 4 b 2 representing two still cancer images. The “2 giant” pieces have been transferred to the host system 1 02, the method of which is to extract and execute the extraction of the machine benefit 1 2 2 0, and obtain the data file 1 1 6 a (from the object 10 8 a), 1 1 6 b, and 11 6 b2 (obtained from object 1 08b), and 1 1 6 c (obtained from object 1 08c). These cases contain digital images in a predetermined common format. Then the data case borrows application 1 2 4 access to manipulate or display in each data file 1 04a, object 1 08a, and associated data and methods 丨 4a and 1 15a

Interaction description. It must be understood that the discussion is not limited to such specific examples, but

2759 V. Description of the invention (5) Applicable to alternative devices and objects. The implementation of one specific example of the system has a bullfight and a method, including those shown in the figure. = The specific examples between the specific examples? It is shown in Figure 2 to assist in understanding the transfer of image differences from the device in this group. The specific examples of learning are d and transfer of shared image data c. According to the invention, the original sensor information is captured. May = The technology starts in step 2 of Figure 2. 4 = The technology is carried out. Second, the image data in the native format, which is based on the original sensor data, is associated with the digital camera. 1 4a f:. —In the specific example, the original sensor data is obtained by removing, color interpolation, shadowing: or proprietary operation processing, such as data in the format of a noisy machine, such as two reductions and scaling up. The sensor information is not indicated by I 2 0 8. In addition, you can skip step 208 '' factory sensor data by any meaningful processing of the project owner. In these cases, possible performance compromises, capacity formats. The processing amount of step 2 08 involves step 212, imaging: pingshu. Then proceed to step 212. The image data ma and the associated I_j4a form an object 1083, which includes a video extraction machine criminal g to execute "V; like Λ method 115a / image method n 5a as a pending form of & · or instruction form for The pre-defined common grid method I 15a of panning image resources by the native format is based on the instruction set of the extraction machine 1220. In most cases, this method is expected to be significantly smaller than the conventional warming module, because the instruction set of the extraction device can optimize the image logic and the reason is to support allocation, such as memory scheduling, which can be built on the extraction machine 120 and Not built on this method. After the imaging device has formed an image object 108a, the operation in Figure 2 continues to transfer the object to the host system. This can be transferred using well-known communication media

Page 9 442759 V. Description of the invention (6) Image data, such as network keying, or using a removable storage device 1 1 〇, such as a non-volatile memory card.

The interface between the communication medium and the extraction machine 120 of the host system is the object driver 122. The object driver 122 communicates with the appropriate hardware and software (including the operating system) of the host system 102 to make the object 108a available for use in a pick-up machine. The object driver starts the extraction machine under the local OS and finally causes it to execute the method. In a specific example, the object drive 1 2 2 includes a well-known body drive for retrieving a file configured according to a tagged image file format (TIFF). Ti FF is a common format used to store bitmap images on PCs and Macintosh computers. It can be used to implement objects 1 0 8 a »Other implements 1 0 8 a are widely used, and associated drives 1 2 2 can be used as Cognitive use of the industry. J a loader (not shown in the figure) loads the method into the extraction machine and sends the-pointer to the image data. The loader also recognizes certain areas of the image object, such as the number of images and iteratively initializes the extraction machine to process each image, as shown in step 22G. This causes the image data to be translated and processed into a common format of the image data. The translated data is usually placed in a separate data coffin, which can be stored in a large-capacity storage device such as a hard disk drive. This information 彳 § case 1 1 6 a can then be further processed by the application 丨 2 4 into a predetermined final form. A specific example of the present invention has been described from the viewpoint of the overall system, and FIGS. 3 and 4 illustrate that other specific examples of the present invention relate to the imaging device 104a and the host system 102. Figure 3 is the host processing system! 〇 2-a specific example, including software and hardware configured to receive and process shirt-like objects 108a. The specific embodiment focuses on the bus 332 coupled to the processor 334. System 102 can be custom-built with Intel processing

Page 10

442 75 S V. Description of the invention (7) The personal computer with the graphical interface and operating system of the device and Microsoft Windows. Although the operating system software 3 4 0, the object driver 1 2 2 and the extraction machine 1 2 0 are shown as being separately loaded in the memory 328, the industry knows that these software components can occasionally be combined and distributed in whole or in part in the memory 328 , Mass storage devices 3 3 8 (such as magnetic disks), and portable memory such as optical disks (not shown). The processor 334 executes instructions stored in a memory 328 (such as random access memory (RAM)) and a large-capacity device 338 = the memory 328 is any machine-readable medium such as a semiconductor integrated circuit, which is executed by the processor 3 34 Instructions can be loaded at any time, so that each configuration step of the host processing system 102 is received for processing objects from different imaging devices and storage devices. The host processing system 1 2 0 also contains a communication interface 3 1 6 which conforms to any of the conventional computer peripheral bus standards, including RS-232 serial, universal serial bus (USB), and I EEE standards 1 3 9 4-1 9 9 5. In addition to communicating over physical lines, it is expected that wireless communication media such as infrared frequencies or radio frequencies may be used for data transmission. As for communication media alternatives, it is expected to include a variety of different removable storage devices, including the Personal Computer Memory Card International Association (PCMC IA), Intel's Flash Micro Cards, and any other non-volatile storage suitable for transmitting digital image data medium. The host processing system 102 of FIG. 3 also has a user display device interface 342 such as a graphics adapter card. The card can interface with any conventional display device including a cathode ray tube (CRT), a liquid crystal display (LCD), and any other suitable viewing interface. Video View > Image Display Technology. Of course, the host system 102 also has other interfaces (not shown) to perform additional 丨 / 〇 (such as a network interface controller).

^ 42759 V. Description of the invention (8) The host part of the specific example of the system of the present invention has been discussed. Fig. 4 illustrates that the imaging device 1 0 4a includes its hardware and software components. Again, the specific example of FIG. 4 is based on a bus 43 2 ′ processor 4 24 and a memory 428 coupled thereto. The alternative here is a microprocessor which replaces the processor 424 and the memory 428. § The memory can be a machine-readable medium such as a semiconductor integrated circuit or a non-volatile semiconductor memory such as a read-only memory (ROM). The memory 4 2 8 is loaded with instructions to be executed by the processor 424. The instructions integrate the image data with the method 1153 as part of the image object 108 (refer to FIG. +1). These instructions The merge instruction is shown in Figure 4 as the object forming software 412. # 像 方法 Π 5a is usually developed as a separate part of the software, which will be described in detail later. Production 5: 8 also includes the optional optional signal processing software 408, 1H software. Used for:-The original sensor data of the next image sensor 4 04 is set at the local machine. The image processing logic circuit 436 also includes-imaging is a complementary image money 4 of ^, which is A specific 7 ld metal supplement emulsion semiconductor (CM0s) actively like Xin senses crying. The software to access the interface 4U and 420 is used to form a processor = processor 424 to execute information. The interface 416 to the communication medium can be like a farmer In the interface, the standard is as described above for the host system, electrical, and peripheral area. The standard can be a J-type mobile storage device, such as a flash: II: Management 'interface 420 to store and transfer image objects 108a ^ 1 & The specific example of the stealth card used for non-volatile Bennumin is shown in Figure 4. Processor and transfer of image objects 108 & formed soil in software to host system

Page 12 412759 V. Description of Invention (9) 102. The alternative is mainly a hardware-based architecture based on gate arrays and application-specific circuits (ASICs), which perform the same functions as previously created in moving image objects 108a. Extraction machines and image objects The remainder of this disclosure discusses techniques for performing image objects 08a and extraction 120, including a set of instructions that can be used to form the image 115a. Remains As discussed earlier, the object 108a can be tagged with an image file / electronic photography (EP) structure or file implementation. The TIFF file includes two " M headers, which point to the image data U4a and the associated method 115 / location. The files and their file headers can be used by the object drive, 22, 1FF, and 120. j Tiantu take the opportunity to develop the imaging method 11 5a The procedure starts with the imaging device 1Q4a comparing the local format and the common format. Developed a kind of algorithm Image data 11 4a from native format to common format. Such algorithms can be developed by the industry. Then the algorithm uses a high-level process such as (: and / or low-level extraction machine 丨 2 0 instructions set in the program factory, detailed: then test the program and selectively optimize the program's heart code into bit π group code It means low-order processing. This becomes the pre-g two method 1 1 5a. Then the method is stored on the outside of the imaging device 1 = memory (for example, memory 428) as $ 10 (see Figure 4). Take part of the autumn body 412. The image object 108a residing in the imaging device 10a is managed in a portable use. It is quite an extra task. It is a question called "such as a digital camera, stored here."

O: \ 59 \ 59455.PTD 42 75 9 V. Description of invention (10) must be a prerequisite. The additional job management is related to the following: G) the processing volume or complexity required to translate the image data 11 4 a from a native format into a common format 'and (2) the data 1 1 4 a in its native format size. By choosing a native format whose computational aspect is close to a pre-defined common format, the imaging method 1 1a becomes a relatively short form of simple math or data movement instructions. Compared to the above, if the image data 1 1 4a needs to be decompressed from a highly compressed native format (such as JPEG or fractai technology) to a common f-type ', the method 11 5a may become quite large. In the examples described below, any dispute over the storage space obtained by compressing the image data is simulated by a more complex and lengthy method of decompressing the image data user ^. In order to solve the storage space problem, the instruction set of the extraction machine can be highly optimized. Various methods are based on the instruction set. Example I round = buffer (receiving image data 11 4a) and output buffer (. Storage The position of the data after translation in the common format) can be used by the extraction machine 120 by the instruction description. The address calculation and most of the data movement can be performed by white instead of specific instructions of the imaging method 115a. As necessary, = 0 = ^ is focused on the algorithm required to translate the associated image data 4a into a common format. In addition to optimizing the instruction set to maintain the simplification of the image method, 5a, Ren = "the method of retaining the external storage area of the imaging device is associated with multiple image data sets.): Method 1 a single copy of 15a, as shown in Figure H. Object 1. 813. In this way, a single image data 10813 contains a plurality of images in the image data 1141) 1 and 114] 32, which actually becomes similar to the conventional negative roll film, and the execution method here is roughly similar to developing the negative film.

Page 14 442759 V. Description of the invention (u) In addition to storage space, another concern of imaging device manufacturers is the protection of intellectual property in image processing methods. This processing is performed on the original sensor data to improve the final image quality, such as interpolated color filter array patterns, r correction, and white balance algorithms. Instead of filling the method 11 5a with such images, the processing algorithm is filled, so when the object 丨 〇8 a is transferred to the host system 〇 02%, these algorithms may be exposed to possible competitors, the imaging device 1 〇 4 a Manufacturers hope that key image processing functions can be ordered inside the imaging device. As a result, the image data 1 i 4a is a strongly processed version of the original sensor data. Taking extreme considerations, manufacturers who hope to protect the image processing technology can embed all the processing inside the imaging device. As a result, the image data q H4a is common. In this case, the corresponding method [15a] does not need to perform any task = job ', only the data 1 1 4 a needs to be copied to the host system. 〇2 The output buffer is better at the other extreme, the image inside the imaging device 〇 04 a Processing can be kept to an absolute minimum. No original processing of the original sensor data is performed before the image data is encapsulated into an object. Method 5a defines all the processing required to achieve the common format, and the extraction machine 12 of the host system performs all processing to achieve the image data in the common format. Virtual machine architecture. An example of a machine 12 is a virtual machine. A virtual machine is a self-contained operating environment that behaves as if the image is executed on a part of the computer ^) has been a. The virtual device assists in shielding the host system from contacting any viruses present in the receiving object = 0 8 a. The virtual machine has an instruction set, a register set, and a memory sub-mode that is independent of the hardware platform of the host processing system 102. Wu Ke with well-known compilers, programming languages, and related tools 1 "

Page 15 442759

Page 16 442 75 9 V. Description of the invention (13). In / when / case) to simplify the program planning and reduce objects as much as possible. 〇8 a flow control additional operation management ^ flexible inspection is also a type change Features required when projecting to another type of variable. Example Virtual Machine-Virtual Image Processor A specific example of the virtual machine may be the following virtual image processor (v 丨 p) β In this specific example, HP does not expect traditional I / O support, nor does it anticipate user's intervention. V IP exists purely for processing image data. 4 a is used as input in the native format, and image data is generated as output in the common format. The memory resources required for this task are usually provided by the host system i 0 2, so the programmer of the image object 108 a does not need to authenticate the memory resource γ-how to provide it. After ^ = description is obviously easy to understand, ν [ρ instruction set is very similar to the conventional hardware ^ ^ instruction set. Part of the reason is that it is desirable to reduce the VTP θ of 5a to ^ and Λ and the reason is to emphasize that most of the image data n 4a is mathematically processed. The operations required for m 5 collection license image processing are specified in the smallest possible slice factor, such as the 'w control currency' instruction, which can be greatly simplified compared to conventional microprocessors.

: Reason, phase: A large number of related complexities can be built into the VI so that an instruction can touch the table ^ earlier. In addition, VIP refers to the number of operations V that are too complex for the microprocessor processor to perform. The number Λ is equal to two, and it is far more than the hard γ column data movement instructions with discrete :: The heat demand is decomposed into a limited physical register set. Instead, it is calculated = two use-groups are present and solved by VIP. ; The learning expression is directly in the image method. The memory is then expressed in the reversed light table, which can be used as the expression of the target indeterminate (RPN). So the calculation

Page 17 442 75 9 V. Description of the invention (14) The element can be directly encoded into the image method 1 1 5 a as a representation component instead of a machine instruction = principle-managed projection and automatic conversion can be seen in any expression. This specific example covers 7 kinds of program control instructions and 5 kinds of data operation control instructions: Instructions. To transfer program control. If to conditional transfer program control, select the program control with index and transfer call transfer program control to the next program return. Return to the program control, exit the program, exit, end the normal execution of the VIP program, and end the abnormal execution of the VIP program. Push the basic unit data variables to the data push stack and jump out from the data stack. Jump out of the basic unit data variables. Generate new data components for program planning. Destroy and release the formed data components. Program control instructions are used to guide the program execution flow. Unlike conventional microprocessors, there are not a large number of conditional branch instructions. And > can directly support index jump (corresponding to the switching program of the C programming language). The target address size is variable with the total size of method i 1 5 a. The data movement instruction is an instruction for actually performing a pan and processing image work. In a specific example, there are no instructions to directly perform arithmetic or logical operations, the reason is

Page 18 U2 7) Bing 9 _ 'A, description of the invention (15) These operations are embedded in the form of the operator associated with each instruction. Also, unlike conventional microprocessors, the operand list of an instruction can have a variable number of operands. The push instruction pushes a data variable onto the data stack. Note data Stacking is different from program flow control stacking. The push instruction is only used to push individual basic unit data types onto the data stack. In other words, it cannot be used to push the entire array onto the stack. The format of the push command is operation ma, followed by the data variable indicator *, which indicates that the data will be pushed to the stack i °. The type and size of the data variable will become the operand. The jump instruction jumps out a data variable from the data stack. The associated data variables need not be of the same type as the data previously pushed onto the stack. If the variable type does not match the existing data type on the top of the stack, you can manage the type projection for automatic type projection according to the same rules as the high-level programming language used to define the method. The move instruction moves data from the source to the destination, and selectively evaluates expressions encoded in the source. In other words, the source need not be a single variable but can be any complex table that is encoded using reverse polishing notation (RPN). The generate command dynamically generates a new data variable and commands a reference value for the variable. The resulting variable can be a single basic unit data type or an array of basic unit data variables. The destroy command releases the memory allocated by the generate command to each data ^ variable. The high-level programming language compiler used to define method 1 1 5 a can be used to generate and destroy instructions for dynamic memory management of automatic variables used in the method 1 15 a subprogram. This is different from the normal practice of using the C programming language which generates automatic variables by stacking. The reason is that VIP is not used as a physical microprocessor and therefore does not contain hardware stacks.

Page 19 4 4275 9 V. Description of the invention (16) VIP high-level language Because the VIP instruction set optimizes the virtual implementation of the image processor, low-level programming languages such as VIP combination languages are rarely similar to the combination languages of conventional microprocessors . It is practically impossible to implement a high-level language such as C or (: ++ full version used to define Method 1 5 a, due to the limitations specific to the aforementioned imaging optimization. Therefore, a suitable compromise is a small set of C language. Use this The VIP program written by the small set C does not include linking steps. Instead, it is compiled into a single file, or it includes a series of files ('eventually becomes a single file). The type of licensed data is declared to be the type used for the aforementioned instruction set. The indicator is ιΤ ”specific type (ie gapless indicator). Indicators to indicators can allow but not more than double pointing. Multidimensional arrays are like the conventional C. Type def and structure announcer behave as expected = complex data types such as The structure can be used as an array element. Several known functions, such as bit blocking, are generally meaningless in terms of the specific content of the VIP, so bit fields are not allowed for simplicity. · Type Controller_ V IP License It is a type controller, but the volatile controller is meaningless because there are no variables for any processing other than the VIP program (such as method 1 1 5 a). Examples of processing complex data types If the structure technology is not provided in the method μ 1 1 5 a, it is instead referenced by the structure definition of the compiled VIP code. This is the key difference between VIP and Xi microprocessor. For example, the structure is double true_part;

Page 20 V. Description of the invention (17) Double hypothesis _ part;} complex [1 0]; short index = 1; complex [index]. True_part = 〇. 〇; The specified statements shown in the section are broken down into a series of instructions to calculate the structural column address based on the known eccentricity of its basic bit address. However, the use of ν IP and the direct result of the wireless address space is because each data item is assigned a unique ID number. The data items may or may not be adjacent to the memory ^ VI P resolves the complex data type reference into a single ID number, which can uniquely distinguish the predetermined variable = 1 V I P can distinguish the ID number from the array index, so it will not be confused. The previous example is a structure array, and each structure contains two data items. After compilation, it can be expressed as: double real_part [1 0]; double hypothetical_part [1 0];

G results obtained equivalent specified program: true_part [indicator] = 0. 0; in this way, the compiler can decompose the structure definition and reference into basic units

Page 21 4 42 75 9 V. Description of the invention (18) and compatible with the storage scheme used by VIP. Storage Types As discussed earlier, the structure of complex data types is contained in the data rather than in the executable code, and the access of the f data is resolved by the reference rather than calculated by the address. There are two exceptions to the specific VIP example: the input data buffer is used to receive image data 1 1 4a, and the round-out data buffer is used to hold the organized image data = there is a continuous byte array, the size of which is defined in the source file When compiled into method 1 1 5 a. The difference between other storage types is a storage class which is not covered by Standard C. There is an input buffer and an output buffer. Any variable declared with an output class of output i_ :) refers to the input buffer. Any variable declared with an output storage class refers to the output buffer. All types defined using one of the storage classes are mapped to individual memory arrays in terms of their starting point. Before method 115a is used to control VIP, two special variables can be set by the loader: input_size and output-size. As such, the byte length of an individual buffer is defined and can be used without first being declared in method 115a (which is defined as valid). In this regard, it is similar to argc and argv as defined by standard C, which has no effect on V I P (it has no purpose). Using these storage classes and special variables, the programmer can map any data type to the input or output buffer. {歹 ^ eg: (. ”Input unsigned integer original—data [input_size / 4]; // announce that the integer array represents the entire original image from the image data 1 1 4 a

Page 22 442759 V. Description of the invention (19) image; input floating original data [input_size / 4]; // announced that the floating array represents the entire original image from the image data 1 1 4a. The variables original_data [η '] and original_fdata [η] represent the same data, but are interpreted differently. The single method 1 1 5 a can process multiple images with different sizes by causing the loader to initialize the input and output buffers Γ) and the size variables of each image. The simplest and feasible image method is to copy directly from the input buffer to the output buffer, because the imaging device 104a generates images in a common format. For example: main () {input byte original [input_size]; — output byte complete [output_size]; complete = original;} wide. Note that the designation statement is similar to the C ++ category designation syntax. This point deviates from the standard C. J syntax and implies parallel processing. Array Any type of array that is used to specify a statement or does not include an explicit indicator notation indicates that the operation belt should be thrown to the overall array, and there is no data phase between the array elements.

Article 23 I V. Description of Invention (20) Relevance. Consider the following code snippet: short X [2 5 6]; short y [2 5 6]; X = y * 1. 4 1 4; each element of the y array can be multiplied by the constant 1. 4 1 4 and the result The corresponding element body assigned to the x array. If the array sizes do not match, you can repeat the calculations for the number of elements in the smaller array. It is implied that calculations and designations can be performed in parallel if possible. This statement does not imply a specific index sequence, "equivalent expression is: x [] = y [] * 1.414; this permits parallel processing of multidimensional arrays if necessary. For example: short row = 6 4 0; short row = 4 8 0; unsigned byte image [row] [column]; image [] [12] = 2 5 5;

This code segment assigns the value 2 5 5 to all the elements of the image array column 12. When compiling this type of higher-order code on a conventional microprocessor, the compiler may generate a sequence of machine instructions to understand the loop operations implied in the statement. However, the VIP compiler simply records the loop operations on time and any required related attempts. VIP itself does not repeat the required indicators. So sacrifice

O: \ 59 \ 59455.PTD Page 24 442759_ V. Description of the invention (21) Can reduce the code size stored in method 1 1 5 a. If the performance of the host system 102 in which the VIP is installed is sufficient to perform the required function at an acceptable rate, then this compromise is acceptable = In other words, many specific examples of the invention have been described as using video Object conception improves the common technology of static image and image data. However, it is expected that the image objects and extraction machine technology can be applied to other types of data, which are also cohesive and bounded, such as certain types of audio or video. In these cases, the video method can be replaced by an audio / video encoding decoder for compressing and decompressing audio / video sequences. Of course, the foregoing specific examples of the present invention can be modified in structure and practice. For example, although FIG. 1 illustrates a single object of each imaging or storage device, this is merely for the purpose of simplifying the description of how the present invention works in specific examples. In fact, people in the industry understand that multiple devices can be used to generate multiple objects and can be migrated to the host system as needed to present most of the images generated by each device. In addition, a variety of other alternatives to the computing architecture of Figures 3 and 4 are also possible, which can achieve image capture and image object formation. Therefore, the scope of the present invention is not limited by specific examples, but is defined by the scope of the accompanying patent application and its legal equivalent.

Page 25

Claims (1)

442759 VI. Scope of patent application 1. An object comprising: a machine-readable medium having instructions which, when executed by a processor, cause the following steps to be associated with the first image data and the first method as one of the image objects In part, the first method is performed by an extraction machine to obtain the first post-translation image data based on the first image. — 2. If the object of the scope of patent application item 1, wherein the machine-readable medium further comprises instructions which, when executed by the processor, cause further steps: associating the second image data with the first method as part of the object The first method is executed by an extraction machine to obtain a second image data based on the second image data.) The shifted image data. 3. For the object of the scope of patent application, the machine-readable medium further includes instructions which, when executed by the processor, cause further steps: associating the second image data with the second method as part of the object, the The second method is performed by an extraction machine to obtain the second translated image data based on the second image data. 4. For the item in the scope of patent application, the data after the first translation is in the same format as the first data. 5—A kind of object, including: a wide. A machine-readable medium having instructions that when executed by a processor, the following steps configure a data processing system to receive the first and second objects from the first and second imaging devices, respectively. , These objects have first and second image data and corresponding methods; and Page 26 442 75 9 6. Scope of patent application An extraction machine executes the corresponding method of each object based on the first and first and second translated image data, respectively. The object of item 5, wherein the first and second images are in the same image file format. Obtaining the second image data 6. If the image data after the scope of the patent application is transferred is 7. a method, including: transferring an image object having a first image data associated with the first method to a processing system; and Extraction machine generates image data from Yanxuan image 8. If the scope of patent application transfer has an associated processing system, the first and the extraction machine perform two translations of the image data in the same format. 9. If the scope of patent application is applied, the processing system executes the first method based on the moved image data. The method of item 7, further comprising: a second object of the second image data of the second method; the second image data is in a different format; a second square is formed; and based on the second image data, the first and second The method of item 7 of the image data after translation is the method of item 7, which further includes: translating the second image data associated with the first method to the processing system; the first method produces the second image data based on the second image data. The image data after the first and second translations should be performed by the extraction machine in the same format. 1 0. — An imaging device image sensor is used for memory to store sensor data and the first includes: generating sensor data and an image object, which has a first image data image method for extracting from a Machine execution Page 27 and Raw Material Materials-Materials Guanlaiji 442759 VI. Patent Application Scope The first image data obtained after the first image data is translated. i 1. The imaging device according to claim 10, wherein the first image data is sensor data. 12 · The imaging device as claimed in item 10 of the patent application, which further includes a processor; and a second memory which has instructions that, when executed by the processor, can cause processing of sensor data to become the first video material. 1 3. The imaging device according to item 12 of the patent application, wherein the processing includes the sensor data and executes the image processing method. 14. The imaging device according to item 10 of the patent application, which further includes (including: a logic circuit for processing the sensor data into the first image data. 1 5. The imaging device according to item 14 of the patent application, The logic circuit executes a color interpolation algorithm on the sensor data. 1 6. The imaging device according to item 10 of the patent application scope, further comprising: _ to a communication medium interface for transferring the first image Materials and the first method to a processing system separate from the imaging device * The processing system is equipped with the extraction machine. &Amp; 1 7. The imaging device according to item 10 of the patent application scope, wherein the image object 1 'contains A TIFF file containing the first image data and the first image method. 1 8. The imaging device according to item 10 of the patent application scope, wherein the translated first image data is a device independent bit mapping table ( DIB) one of the formats Page 28 42759 VI. Scope of patent application Part of the image file. 19. The imaging device according to item 10 of the patent application, wherein the first image data and the translated first image data have the same image file format. 2 0. A data processing system comprising: a processor; a memory coupled to the processor and having instructions which, when executed by the processor, cause the following steps to configure the system, which are then received by the first and second imaging devices, respectively The first and second objects, each object has image data and a corresponding method; and an extraction machine executes the corresponding method of each object based on the image data 彳]; corresponding data after translation is obtained. 2 1. If the system of patent application Fanyuan 苐 20, where the translated data is part of the first and second image files with the same image file format. Page 29